File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Memristor-Based Analog Computation and Neural Network Classification with a Dot Product Engine

TitleMemristor-Based Analog Computation and Neural Network Classification with a Dot Product Engine
Authors
Keywordsneuromorphic computing
crossbar arrays
metal oxide
memristor
Issue Date2018
Citation
Advanced Materials, 2018, v. 30, n. 9, article no. 1705914 How to Cite?
Abstract© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Using memristor crossbar arrays to accelerate computations is a promising approach to efficiently implement algorithms in deep neural networks. Early demonstrations, however, are limited to simulations or small-scale problems primarily due to materials and device challenges that limit the size of the memristor crossbar arrays that can be reliably programmed to stable and analog values, which is the focus of the current work. High-precision analog tuning and control of memristor cells across a 128 × 64 array is demonstrated, and the resulting vector matrix multiplication (VMM) computing precision is evaluated. Single-layer neural network inference is performed in these arrays, and the performance compared to a digital approach is assessed. Memristor computing system used here reaches a VMM accuracy equivalent of 6 bits, and an 89.9% recognition accuracy is achieved for the 10k MNIST handwritten digit test set. Forecasts show that with integrated (on chip) and scaled memristors, a computational efficiency greater than 100 trillion operations per second per Watt is possible.
DescriptionAccepted manuscript is available on the publisher website.
Persistent Identifierhttp://hdl.handle.net/10722/286954
ISSN
2023 Impact Factor: 27.4
2023 SCImago Journal Rankings: 9.191
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorHu, Miao-
dc.contributor.authorGraves, Catherine E.-
dc.contributor.authorLi, Can-
dc.contributor.authorLi, Yunning-
dc.contributor.authorGe, Ning-
dc.contributor.authorMontgomery, Eric-
dc.contributor.authorDavila, Noraica-
dc.contributor.authorJiang, Hao-
dc.contributor.authorWilliams, R. Stanley-
dc.contributor.authorYang, J. Joshua-
dc.contributor.authorXia, Qiangfei-
dc.contributor.authorStrachan, John Paul-
dc.date.accessioned2020-09-07T11:46:06Z-
dc.date.available2020-09-07T11:46:06Z-
dc.date.issued2018-
dc.identifier.citationAdvanced Materials, 2018, v. 30, n. 9, article no. 1705914-
dc.identifier.issn0935-9648-
dc.identifier.urihttp://hdl.handle.net/10722/286954-
dc.descriptionAccepted manuscript is available on the publisher website.-
dc.description.abstract© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Using memristor crossbar arrays to accelerate computations is a promising approach to efficiently implement algorithms in deep neural networks. Early demonstrations, however, are limited to simulations or small-scale problems primarily due to materials and device challenges that limit the size of the memristor crossbar arrays that can be reliably programmed to stable and analog values, which is the focus of the current work. High-precision analog tuning and control of memristor cells across a 128 × 64 array is demonstrated, and the resulting vector matrix multiplication (VMM) computing precision is evaluated. Single-layer neural network inference is performed in these arrays, and the performance compared to a digital approach is assessed. Memristor computing system used here reaches a VMM accuracy equivalent of 6 bits, and an 89.9% recognition accuracy is achieved for the 10k MNIST handwritten digit test set. Forecasts show that with integrated (on chip) and scaled memristors, a computational efficiency greater than 100 trillion operations per second per Watt is possible.-
dc.languageeng-
dc.relation.ispartofAdvanced Materials-
dc.subjectneuromorphic computing-
dc.subjectcrossbar arrays-
dc.subjectmetal oxide-
dc.subjectmemristor-
dc.titleMemristor-Based Analog Computation and Neural Network Classification with a Dot Product Engine-
dc.typeArticle-
dc.description.naturelink_to_OA_fulltext-
dc.identifier.doi10.1002/adma.201705914-
dc.identifier.pmid29318659-
dc.identifier.scopuseid_2-s2.0-85040165692-
dc.identifier.volume30-
dc.identifier.issue9-
dc.identifier.spagearticle no. 1705914-
dc.identifier.epagearticle no. 1705914-
dc.identifier.eissn1521-4095-
dc.identifier.isiWOS:000426491600032-
dc.identifier.issnl0935-9648-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats